PCDA-MPAT：物理约束数据增强与多径Transformer的无源物联网室内定位方法

肖善鹏; 范艺晶; 郑师应; 曹艳艳; 王亚敏; 邢飞

doi:10.12263/DZXB.20260456

您当前的位置：

首页 >

文章列表页 >

PCDA-MPAT：物理约束数据增强与多径Transformer的无源物联网室内定位方法

射频计算 | 更新时间：2026-10-16

- PCDA-MPAT：物理约束数据增强与多径Transformer的无源物联网室内定位方法
- PCDA-MPAT: Physically Constrained Data Augmentation and Multipath-Aware Transformer for Ambient IoT Indoor Localization
- 电子学报 2026年54卷第6期页码：2522-2533
- 作者机构：
  
  1.清华大学精密仪器系，北京 100084
  2.中国移动通信研究院，北京 100053
- 作者简介：
  
  肖善鹏男，1976年1月出生于山东省菏泽市。现为中国移动研究院物联网技术与应用研究所所长、高级工程师。主要研究方向为无源物联网、无线感知及无线通信系统等。E-mail: xiaoshanpeng@chinamobile.com
  范艺晶女，1997年1月出生于陕西省宝鸡市。现为中国移动研究院研究员。主要研究方向为无源物联网、无线感知、深度学习等。E-mail: fanyijing@chinamobile.com
  郑师应女，1992年10月出生于四川省宜宾市。现为中国移动研究院技术经理。主要研究方向为无源物联网、无线通信系统等。E-mail: zhengshiying@chinamobile.com
  曹艳艳女，1983年8月出生于江苏省宿迁市。现为中国移动研究院正高级工程师。主要研究方向为物联网体系架构、无源物联网、无线通信系统等。E-mail: caoyanyanwl@chinamobile.com
  王亚敏女，1979年11月出生于山东省菏泽市。现为中国移动研究院正高级工程师。主要研究方向为复杂系统智能优化、无源物联网、物联网等。E-mail: wangyamin@chinamobile.com
  邢飞男，1979年2月出生于辽宁省葫芦岛市。现为清华大学教授。主要研究方向为微小卫星及微小型高精度姿态测量传感器、微米纳米测量与智能微系统技术等。中国电子学会会员编号：E190202752M。E-mail: xingfei@mail.tsinghua.edu.cn
- 基金信息：
  
  国家科技重大专项(2025ZD1301400)
- DOI：10.12263/DZXB.20260456
  中图分类号： TP391;TP183
- 收稿：2026-05-08，
  
  录用：2026-06-01，
  
  纸质出版：2026-06-25
- 稿件说明：
移动端阅览
肖善鹏, 范艺晶, 郑师应, 等. PCDA-MPAT：物理约束数据增强与多径Transformer的无源物联网室内定位方法[J]. 电子学报, 2026, 54(06): 2522-2533.

XIAO Shanpeng, FAN Yijing, ZHENG Shiying, et al. PCDA-MPAT: Physically Constrained Data Augmentation and Multipath-Aware Transformer for Ambient IoT Indoor Localization[J]. Acta Electronica Sinica, 2026, 54(06): 2522-2533.
肖善鹏, 范艺晶, 郑师应, 等. PCDA-MPAT：物理约束数据增强与多径Transformer的无源物联网室内定位方法[J]. 电子学报, 2026, 54(06): 2522-2533. DOI：10.12263/DZXB.20260456

XIAO Shanpeng, FAN Yijing, ZHENG Shiying, et al. PCDA-MPAT: Physically Constrained Data Augmentation and Multipath-Aware Transformer for Ambient IoT Indoor Localization[J]. Acta Electronica Sinica, 2026, 54(06): 2522-2533. DOI：10.12263/DZXB.20260456

摘要

无源物联网（Ambient Internet of Things，Ambient IoT）室内定位技术在智慧仓储、新零售等场景中具有重要应用价值，但室内复杂多径环境会引发两方面突出问题：一是多径遮挡和标签性能差异导致强弱标签样本严重不均衡，在典型服装陈列场景中，弱标签有效样本量仅为强标签的30%左右；二是多径混叠使信号的位置相关特征区分度显著下降，制约了定位精度的提升。为解决上述问题，本文提出一种融合物理约束数据增强与多径感知Transformer（Physically Constrained Data Augmentation and MultiPath-Aware Transformer，PCDA-MPAT）的定位方法。在数据增强层面，本文提出物理约束的数据增强（Physically Constrained Data Augmentation，PCDA）方法，以相位-频率斜率一致性作为物理约束条件，构建带有梯度惩罚的条件Wasserstein生成对抗网络（Wasserstein Generative Adversarial Network with Gradient Penalty，WGAN-GP），生成既匹配实测数据分布又符合无线传播物理特性的弱标签样本，从数据层面有效缓解了强弱标签样本不均衡问题。在模型设计层面，本文构建多径感知Transformer（MultiPath-Aware Transformer，MPAT）定位模型，设计适配准静态多径信道慢衰落特性的时域局部窗口注意力机制，结合时域相对位置编码，捕捉多径特征的时域演化规律，抑制长时域不相关的噪声干扰，增强模型对多径特征的提取与学习能力。在10 m × 10 m的服装陈列场景中开展实测实验，结果表明本文方法的标签级定位准确率达到83.57%，较现有最优基线方法提升5.71个百分点。消融实验进一步验证了PCDA与MPAT的有效性，其中，PCDA使标签级定位准确率提升4.29个百分点，MPAT在此基础上进一步提升2.14个百分点。上述结果验证了本文方法在复杂室内多径环境下的有效性，可满足智慧仓储、新零售等场景的定位需求。

Abstract

Ambient internet of things (Ambient IoT) indoor localization technology holds significant application value in scenarios such as smart warehousing and new retail. However

complex indoor multipath environments give rise to two prominent issues: first

severe imbalance between strong and weak tag samples caused by multipath blockage and tag performance variations. In a typical clothing display scenario

the effective sample size for weak tags is only about 30% of that for strong tags. Second

multipath aliasing significantly degrades the distinguishability of location-dependent signal features

limiting improvements in localization accuracy. To address these issues

this paper proposes a physically constrained data augmentation and multipath-aware transformer (PCDA-MPAT) localization method that integrates physically constrained data augmentation with a multipath-aware Transformer. At the data augmentation level

we propose a physically constrained data augmentation (PCDA) method. Using phase-frequency slope consistency as a physical constraint

we construct a conditional Wasserstein generative adversarial network with gradient penalty (WGAN-GP) to generate weak tag samples that not only match the distribution of measured data but also adhere to the physical laws of wireless propagation

effectively alleviating the imbalance between strong and weak tag samples from the data perspective. At the model design level

we construct a multipath-aware transformer (MPAT) localization model. We design a time-domain local window attention mechanism adapted to the slow-fading characteristics of quasi-static multipath channels

combined with time-domain relative position encoding

to capture the temporal evolution patterns of multipath features

suppress long-term irrelevant noise interference

and enhance the model’s ability to extract and learn multipath features. Experimental results in a 10 m×10 m clothing display scenario demonstrate that the proposed method achieves a tag-level localization accuracy of 83.57%

outperforming the best existing baseline method by 5.71 percentage points. Ablation studies further validate the effectiveness of PCDA and MPAT. PCDA alone improves the tag-level localization accuracy by 4.29 percentage points

and MPAT further increases it by 2.14 percentage points. These results confirm the effectiveness of the proposed method in complex indoor multipath environments

meeting the localization requirements of scenarios such as smart warehousing and new retail.

关键词

Keywords

references

Kaune R . Accuracy studies for TDOA and TOA localization [C ] // 2012 15th International Conference on Information Fusion . Piscataway : IEEE , 2012 : 408 - 415 . DOI: 10.1109/sdf.2012.6327908 http://dx.doi.org/10.1109/sdf.2012.6327908

Dela Cruz R , Rhodette Pedrasa J . RSSI-based localization in an emulated active RFID system [C ] // 2019 IEEE 10th Annual Information Technology, Electronics and Mobile Communication Conference . Piscataway : IEEE , 2019 : 771 - 778 . DOI: 10.1109/iemcon.2019.8936285 http://dx.doi.org/10.1109/iemcon.2019.8936285

Liu Junyang , Li Yuan , Zhang Yulu , et al . Passive IoT localization technology based on SD-PDOA in NLOS and multi-path environments [J ] . Computers , Materials & Continua, 2024 , 80 ( 1 ): 913 - 930 . DOI: 10.32604/cmc.2024.049999 http://dx.doi.org/10.32604/cmc.2024.049999

Ni L M , Liu Yunhao , Lau Y C , et al . LANDMARC: Indoor location sensing using active RFID [C ] // Proceedings of the First IEEE International Conference on Pervasive Computing and Communications , 2003 . Piscataway : IEEE , 2003: 407 - 415 . DOI: 10.1023/b:wine.0000044029.06344.dd http://dx.doi.org/10.1023/b:wine.0000044029.06344.dd

Hatem E , Abou-Chakra S , Colin E , et al . A reliable propagation channel model for a better RFID-based indoor localization system [C ] // 2022 4th IEEE Middle East and North Africa Communications Conference . Piscataway : IEEE , 2022 : 18 - 23 . DOI: 10.1109/menacomm57252.2022.9998285 http://dx.doi.org/10.1109/menacomm57252.2022.9998285

Nanni F M C , Marrocco G . Static and dynamic fingerprint of RFID devices [C ] // 2024 9th International Conference on Smart and Sustainable Technologies . Piscataway : IEEE , 2024 : 1 - 3 . DOI: 10.23919/splitech61897.2024.10612662 http://dx.doi.org/10.23919/splitech61897.2024.10612662

Xie Yaqin , Wang Xiaoli , Lin Peng , et al . Relative positioning of the inspection robot based on RFID tag array in complex GNSS-denied environments [J ] . IEEE Transactions on Instrumentation and Measurement , 2024 , 73 : 5503214 . DOI: 10.1109/tim.2024.3412203 http://dx.doi.org/10.1109/tim.2024.3412203

Pan Zemin , Chen Cong , Yu Yishuai , et al . Research on SVM-based RFID indoor location [C ] // 2023 International Conference on Intelligent Sensing and Industrial Automation . New York : ACM , 2023 : 1 - 6 . DOI: 10.1145/3632314.3632337 http://dx.doi.org/10.1145/3632314.3632337

Jiang Chaoyu , Wang Boyu , Tai T O , et al . An intelligent occupancy detection system for smart tourism based on RFID passive tag antenna array and random forest [C ] // 2023 IEEE 13th International Conference on RFID Technology and Applications . Piscataway : IEEE , 2023 : 83 - 86 . DOI: 10.1109/rfid-ta58140.2023.10290682 http://dx.doi.org/10.1109/rfid-ta58140.2023.10290682

He Lin , Fang Zirui , He Yigang , et al . DFRP: Multipath robust dual-feature RFID positioning using phase and RSSI for food-industry warehousing [J ] . IEEE Transactions on Instrumentation and Measurement , 2026 , 75 : 8002911 . DOI: 10.1109/tim.2026.3677935 http://dx.doi.org/10.1109/tim.2026.3677935

Lv Mingfeng , Wang Zhijing , Huang Yibin . DR-ILS: High-precision RFID indoor localization via differential RSSI and iterated local search [C ] // 2026 International Conference on Communication Networks and Machine Learning . Piscataway : IEEE , 2026 : 605 - 609 . DOI: 10.1109/cnml68938.2026.11452997 http://dx.doi.org/10.1109/cnml68938.2026.11452997

Amoah B , Zhang Jian , Mao Shiwen , et al . EchoNet: Breaking through multipath barriers for high-precision RFID-IoT sensing [C ] // 2025 IEEE Middle East Conference on Communications and Networking . Piscataway : IEEE , 2025 : 1 - 6 . DOI: 10.1109/mecom67453.2025.11439482 http://dx.doi.org/10.1109/mecom67453.2025.11439482

Yong Yunfen , Tan C K , Tan I K T . SMOTE for Wi-Fi fingerprint construction in indoor positioning systems [C ] // 2021 IEEE International Performance, Computing, and Communications Conference . Piscataway : IEEE , 2021 : 1 - 6 . DOI: 10.1109/ipccc51483.2021.9679418 http://dx.doi.org/10.1109/ipccc51483.2021.9679418

Wu Jialing , Li Wanyi , Wu Yilin , et al . Wasserstein generative adversarial network with gradient penalty for handwritten digit generation [C ] // 2024 International Conference on Intelligent Robotics and Automatic Control . Piscataway : IEEE , 2024 : 375 - 379 . DOI: 10.1109/irac63143.2024.10871435 http://dx.doi.org/10.1109/irac63143.2024.10871435

Faseth T , Winkler M , Arthaber H , et al . The influence of multipath propagation on phase-based narrowband positioning principles in UHF RFID [C ] // 2011 IEEE-APS Topical Conference on Antennas and Propagation in Wireless Communications . Piscataway : IEEE , 2011 : 1144 - 1147 . DOI: 10.1109/apwc.2011.6046829 http://dx.doi.org/10.1109/apwc.2011.6046829

Tse D , Viswanath P . Fundamentals of wireless communication [M ] . Cambridge : Cambridge University Press , 2005 . DOI: 10.1017/cbo9780511807213 http://dx.doi.org/10.1017/cbo9780511807213

Vaswani A , Shazeer N , Parmar N , et al . Attention is all you need [PP/OL ] . V7. arXiv ( 2023-08-02 )[ 2026-05-08 ] . https://arxiv.org/abs/1706.03762 https://arxiv.org/abs/1706.03762 . DOI: 10.65215/ctdc8e75 http://dx.doi.org/10.65215/ctdc8e75

Yang Songzhen , Jin Meng , He Yuan , et al . RF-prism: Versatile RFID-based sensing through phase disentangling [C ] // 2021 IEEE 41st International Conference on Distributed Computing Systems . Piscataway : IEEE , 2021 : 1053 - 1063 . DOI: 10.1109/icdcs51616.2021.00104 http://dx.doi.org/10.1109/icdcs51616.2021.00104

Bello P . Characterization of randomly time-variant linear channels [J ] . IEEE Transactions on Communications Systems , 1963 , 11 ( 4 ): 360 - 393 . DOI: 10.1109/tcom.1963.1088793 http://dx.doi.org/10.1109/tcom.1963.1088793

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于深度学习的单帧图像超分辨率重建综述

小样本条件下基于数据增强和WACGAN的雷达目标识别算法

面向工业多时间尺度软测量的时序解耦与度量对齐Transformer模型

低轨卫星环境下OTFS跨域嵌入式导频信道估计算法

面向多类型二值分割任务的混合专家自适应统一框架